The comments/feedback is crazy!! Thank you all for your awesome ideas, suggestions, tips, tricks and whatnot. Really awesome to hear so many advices, opinions. I think part 2 of this project will be a "community" video in which I will partly test your ideas. Will be lots of fun. But it will certainly take a while until I get to that. So stay tuned ;-)
I don't understand why you didn't mount the lights under each step and put pressure sensors in the floor at the top and bottom... it's what I did at least.
@@philosoaperit would indeed be the easiest, but that would probably add a bunch of wiring that you can't really hide unless you start drilling and/or creating some slots.
Specially since a considerable part of those end up being debunked as just not showing the tons of fixes behind the scenes, or downright faked to work.
The reason more channels don't show all their failed attempts, is probably mostly about to saving time. As we know - it takes quite a bit of luck for a project to work right away. Troubleshoting and possibly redesigning a few times is usually a part of designing something - which often takes quite a bit of time on it's own. Now if making video documenting each step, redraw scematics to show of every change then edit all those video shorts, the project may take "forever" to get done, as more time will often be spent making the video, than on the project it self. It's just so much easier to just wait with the filmning and editing, until everything is already finished and tested - then make a detailed video how it works with scematics and how to build. But yes, it can still be of interest those who take their time to even show the failed attempts and for those new to electronics, it can be very useful
I found where you have the problem!!!! I also double checked with a electronics technician that has 50+ years in component level troubleshooting and high frequency circuit design who just happens to be my father that my suspicions where correct. He was able to give a solution and explain why it was happening along with why you were getting that specific frequency. Your driver circuit for the LED needs a second filter on the other side of the coil (U6) but before the diode (D3). In the current configuration, with the PWM being fed into there, you are pumping a RLC circuit in a single direction that has no where to release its energy other than ground and the power monitoring pin of the micro-controller (pin PA6 of part U3). Since it is going though a series of diodes, it is reflecting off of them and this is causing a doubling of the frequency. If you run a RLC calculation f = 1 / [2π × √(L × C)], you get 7.341 kHz for with the ideal component values and 14.682 kHz for the doubled, which is within 95% of the ideal component values, so accounting for the variance in the parts and the sub-harmonics we are seeing on the oscilloscope, this is pretty good proof that the hypothesis is correct. To test it further, removing a single unit for testing would be advisable and then replacing filter capacitor C10 with a different value or adding one in parallel to change the value and then checking with an oscilloscope might be a possible way to check for it. It is also possible that the oscillating circuit is internal to the LED driver IC, so adding this cap may do nothing, but adding the filter at the location that I specify as the solution is going to still work as that is the correct filter location in the circuit. You are going to need to put rejection filters between U6 and D3 that only pass DC and the fundamental frequency for your PWM signal. You could also get away with a low pass filter if the fundamental frequency for the PWM is low enough. I would suggest changing the PWM frequency to a lower value so that you can just use the low pass filter as high frequency noise is easiest to filter out. The datasheet for the micro-controller shows how to do this. As long as the resulting frequency is at least 10% below 7 kHz it should get rid of the noise. Assuming you are on the default settings for it, you should be well below that, but double check. Now as to the why on why this was causing so many problems with the micro-controller. The default clock speed of the micro controller might be set to 16 kHz (+/- 3% over voltage range and +/- 2% over temp. range) by default, so, if you have a signal on the other end of the PWM output pin that is close to that and really noisy while being unstable, like you have, and that pin isn't optically isolated (which this micro-controller doesn't have optical isolation on its I/O pins and problems like this are why you want optically isolated I/O pins) the interference can be feeding back through the circuit and causing problems with the PWM circuit. 97% of 16 kHz is 15.52 kHz, which is close enough for some signal coupling, so it can feed back into the circuit and cause the clock for the PWM to start having errors that the processor will have to correct, which will cause all sorts of strange behavior. I would change the internal clock that the PWM is using to the 20 kHz mode that the micro-controller has the option of using to just completely take it out of using a similar frequency. Make sure that the fundamental frequency that gets produced by the PWM still stays below the 7 kHz range using the equations given in the datasheet for how resolution and clock selection and clock type result in the output PWM signal's fundamental frequency. Just putting the filter in will probably fix most of the noise, but lowest noise possible is the best thing to do when dealing with micro-controllers. I would suggest in future only using micro-controllers with optically isolated outputs and always make sure that if you have a cap and coil in series in an oscillating circuit with diodes that you have a filter for both directions that the coil can discharge unless you are trying to make it oscillate or build a higher voltage.
U1 is the LED driver at 7:30. The circuit he used is exactly the same as Diode's reference design, it is just drawn a bit weird. The chip itself is a bit weird for using high-side sensing instead of low-side. Aside from that, it is a completely ordinary buck regulator. The only thing "missing" is an optional filter capacitor across the LEDs at the regulator, which may help with EMI when the LED current loop is taking a ~35cm trip around the board.
@@teardowndan5364 I know it is buck converter and it is going to have a ring to it there unless you add a filter where I said to put it, they all do and that is the frequency of that specific circuit's oscillation. The problem he is getting is that the frequency it is ringing at is really close to the clock source the PWM circuit is using in the micro-controller. This is a problem because the micro controller doesn't have optically isolated I/O pins, so any high frequency signal that gets on one of those pins that is close to one of the frequencies the controller IC uses, it will propagate into it and start causing errors intermittently. He might be able to get away with just changing the clock value in the controller from 16 kHz to 20 kHz. If not, he needs to filter off that noise. A buck converter circuit is noisy. Connecting it to a digital circuit or a circuit with active components can result in weird behaviors since you are going to get noise from the buck converter that is in the kHz to MHz range, both of which are in the RF range where they start treating everything as active components, even the case of the circuit, so when it gets into an IC, things get weird. I would use a different power source for driving the LEDs in this application next time or isolate the power supply from the micro controller, like with optical isolation.
@coolmonkey619 2 year degree at a technical college will do it. A BS or BA in electrical engineering will also cover it. You are probably going to need a 3.0 or higher GPA or at least paying attention to the topics of power supply supplies, digital circuits, and high-frequency circuits. Just for this problem, having a basic understanding of IC internal design and clock generating is also needed for this as well. Then you also need to understand micro controllers along with microprocessor error correction to see why the problem was so weird. I only confirmed my suspension about the problem with my father, who is a retired microwave telecom tech and designer because I could find the multiplier in the circuit that was giving the 15 kHz, but that is mainly because I would never have ran across this problem since my instructors warned me about the pitfalls of noisy power supplies and digital circuits along with micro-controllers needing the optical isolation or you can risk noise backfeeding into the I/Os or a transient coming in and frying the micro-controller.
@@KnightsWithoutATable If Diodes doesn't include any additional filtering, I am pretty sure it is because its engineers who work on this stuff full-time have done every analysis prescribed by best practices and have concluded that it is perfectly fine as-is for the IC's intended application and following applicable layout guidelines, which usually includes keeping all high-frequency stuff tightly packed, which it isn't when you have two LEDs in series spaced ~15cm apart and the traces in-between acting as an antenna. Just had a look at his PCB layout. The LED trace routing is an EMI disaster. The current return trace needs to be parallel with the source trace like a differential pair, not routed along opposite board edges! And those power traces are also running parallel to the data traces on the other board layer. Maximum EMI.
@@greatscottlab a tube of acrylic may help with channeling the emitting side of the diode, which can then be shielded with heatshrink. A 3D printed bracket to hold this in place sounds perfect. 🙂Perhaps the V2 PCB's could have some holes in them to allow for locating and/or installing these brackets or features, if only for those who need it getting the inspiration.
To be honest this was one of the most eye-opening of your videos to me. Seeing such a theoretically simple and trivial project experience issues due to noise and other things.
If you redesign the boards, it might be easier to use connectors to slide them together, soldering them all together will have to many failure points and you cant take it apart to diag and repair.
@@greatscottlab I was thinking a surface mount, right angle, pin header like M22-5330405 (found on digikey) for low profile easy connections. I’m sure you can find something like it cheaper and quicker on JLCPCB, but their connector page wouldn’t load for me.
@@greatscottlab also make them so that you can disconnect them without sliding out everything, maybe even make them 2mm smaller and glue them into place so that you can take them out easily
Honestly I think that idea could be commercialized pretty easily. If they are snap together all-in-one units that a do-it-yourselfer could easily accomplish... that would be a great selling product.
The design of the data lines won't work as indended as described at 5:12, there should be one diode for each OUT pad, not one for both. In your current design, every other board is commoned together via the OUT-Line of the board in between: Board A In2↔Board B OUT1/OUT2 ↔Board C IN1/IN2 ↔Board D OUT1/OUT2↔Board E IN1/IN2... Most likely that's why all the boards were lit in most of your initial trouble shooting. The boards at the top of the stair (which later you found missbehaving) pulled all the other inputs low, until you introduced smaller pull-ups until the voltage drop was enougth to be registered as low.
I noticed that too. His outputs are connecting to outputs and his inputs are connecting to inputs in his drawing. If it's the same on the actual PCB, then that would explain a lot
@greatscottlab LED Lighting Electronics Engineer here. I reviewed your project files and a few things: 1) put 0.1uF and 10uF cap on the input and output of your LED switcher. You actually need a high freq cap and lower freq cap to cover the freq range for input and output ripple 2) I like the 4-layer idea, but do 2 internal layers of ground, and only route on top and bottom. Also, make sure to sink a ground via right next to every signal line via (this will help the signal keep good ground reference and force the inductive coupling on the internal layers to follow a set path. Try to reduce the number of vias per net as much as possible (don't want added inductance) 3) Make sure to look at the recommended switcher layouts and make your switching nets small, and DO NOT let them go onto other layers (don't want to have inductive coupling between layers). Also, try and get ground between traces so you don't get cross talk. At the lengths of your PCBs, I'd expect you could get Volts worth of cross talk, even at sub 100MHz freq. 4) I'd add a common mode choke to the input of every PCB and a Pi-Filter too. It may be overkill, but for a few extra bucks you grantee each board isn't getting noise on the input. Make sure to route this all on the same layer as the connection between boards (so top layer for you). The shear length of the boards all ganged together is working against you and stray fields from surrounding power lines and appliances/electronics could get onto the boards and cause some weird behavior. Better to go all out and find you you don't need it then to find out you do need it haha. I highly recommend "philslab" on youtube as has some great videos giving best practices. The whole FEDEVEL group is very helpful for PCB design for EMI / EMC. Wish you the best of luck man! I love your videos and love the idea ;) Keep it up man!
Frugal Engineer : This can be done with two proximity sensors, the 1st at the beginning of the staircase and the 2nd at the end. Keep the entire LED strip on until you pass the ladder. Use reliable MCU and LED drivers. As per my understanding, you don't want to manually turn on/off the house lights, and you don't want big bulbs to be lit
@@albertoandrescuevashiguera4304 We understand what he wants and that would make for a rather cool effect but seriously there's no need to overcomplicate lights in this way, especially when they're only going to be active for about a minute once or twice per night, I mean it's just for lighting stairs just leave them on all night at the lowest visible brightness...
@@albertoandrescuevashiguera4304 Doesn't make any difference, it serves the purpose, that's the only thing that should matter logically. Keep the LEDs on until you're done, not just the steps, but the entire staircase.
There is a concept humans have they call "aesthetics". A little weird to us, but it mainly means they like things just from their look and feel. More than that: This can differ in every one of them! To us it has a logical benefit, though: It covers the case where someone turns around on the stairs or two people go. Which would be no issue if both sides just trigger a timer, though.
When I made some under cabinet led tape lights integrate with doors and mains power about five years ago, I had a similar story. A "weekend project" spiraled into about 8 work days of effort and required a one year break for "finishing touches". I could not be happier with the results, even thought it needs another minor touch up. Your project is so much more ambitious and I cant wait to see how it turns out!
lol, it was quite funny. just like my wife, 'you are not going to put a solar light up here'. Our porch needed a motion light, and she insisted it not be visible, so I had to hide it all.
You could use the WS2811 IC, it's the one that is in the neopixel, just replace the RGB LEDs with white LEDs, like that you can control 3 white LEDs with one data line. That way removes the need to have a constant-current source (or sink) to drive the LEDs with a PWM (or any switching power regulation), and then also removes the high frequency noise that could be induced in the circuit.
@@greatscottlab I'd just put in the WS2812/13 LEDs that have the IC built in, saves component cost and you can build in a april fools version where the stairs turns itself into a disco just 'because you can' :P
@3zuli @someguy4915 These options would also work, it will also allow to have red night light but I was supposing that @greatscottlab didn't want to have/use RGB LEDs and wanted better white LEDs.
I came accross this brilliant video a couple of weeks ago and absolutely loved it, including the issues and how you try to deal with them. All of a sudden, I don't feel alone anymore :). I found the idea very good, so I had to try it and sourced a few components and knoked up a few boards on the micro cnc and wired everything > I modified ever so slightly the schematic and the code to address some logic issue with the pin In and Out. On one board everything works, but as soon as you connect several boards they start firing up randomly. Like you I spent a bit of time trying to figure out what happened. My conclusion was that it was nothing to do with the noise on the in/out line. In fact the issue is the variance in the response from the VCNL4200 sensor from one unit to the next and also within the same unit!. So I finally got it to work by measuring a "reference threshold" in the setup of the code (so I might have different value for each sensor) and also by taking the average of 11 samples for each reading of the proxy sensor. That completely fixed my issue! (I think ;) ) If you are interrested in the code or schematic I can of course share it. Thanks again for your great work!
I love that you show what didn't work. I'm still at the beginning and my first thought would be to illuminate the whole staircase dimly with a few presence detectors (including beginning & end) or use buttons for simplicity and use a timer to turn it off after a suitable amount of time. Illuminating each few steps individually seems like an anti-feature feature because we literally lookahead to plan our steps in advance.
maybe if he tried it with light diffusion material with narrow holes just for the sensor's diodes, I think its small enough to disappear when looking from a distance.
That's what I was thinking too - however a tube may not solve the issue as there may still be light reflected from other surfaces / feet. It might be an option, since you have comms between the boards, to activate the sensors one at a time so it effectively scans the stairs. Then each sensor will only receive it's own light.
For the noise problem coming from the switching power supply, I suggest you to use shielded inductors. And maybe, you can also try to connect to ground the aluminium bar. It can act as a shield. I also suggest you to use some sort of connectors between the modules, instead of soldering them together, so if a module need to be replaced, you can do it easily.
Oh man I LOVE the "we just got this staircase, you will not damage it in any way". Any tech man needs a woman like this to keep things sane. My girl does the same. You girl is a keeper :)
You have noise on the Input Pin (PA7) Add a 100nF capacitor between PA7 and GND. Place the capacitor close to the ATtiny402. When using long wires, in your case long traces, you made a Antenna. All input pins should have Capacitors on them, but the higher the capacitance, the lower the switching speed, so you'll have a lower transmission rate
Can you maybe do a video about "Choosing the right microcontroller for your project"? Or generally explain how you search for a suitable microcontroller
It pretty much breaks down to the following: 1. Do you already own the expensive programmer for a particular brand of micro controller. (start looking here and use one unless there are litterally none that meet the other following requirements) 2. Consider the number of general I/O pins you need and if you need any special features like a DAC or audio peripherals (essentially anything you want built into the microcontroller and would rather not have it on a seperate IC) 3. Is the microcontroller fast enough / has enough flash memory for your program. If you're using Arduino consider a micro controller with at least 8KB of Flash so you have a good amount of room for your actual program. Also make sure it's fast enough to do what you need it to do if you're using an ISR for example to do something. Those are pretty much the basics for choosing a micro controller for a project.
My solution: Up and down is one HLK-LD2410, both connected to ESP8266. The bottom one has one pin connected to LED strip with WS2812 LEDs on the whole path of the stairs. Once motion is detected near the bottom one, the LEDs will light up from bottom to top. Once the motion is detected on the top one, it sends the signal to the bottom one and the stairs will light from top to bottom in "usual" walking pace. Works great, keeps the light on for next 5 seconds, to allow kids to, for example, find the entrance to bathroom, or to the next staircase, then the lights "go out" in the same way the came in. Works great, the whole solution costed less than $30, if I do not include the aluminum LED channels - those were the most expensive part of the project in the end :) And as this is radar, I have no problem with transparency of the plastic ;)
You are indeed making me feel my age. When I was young, from birth to age 12, we lived on what was called the "home place' an old house built by grandfather in 1910. While REA came through and put a high line pole in our yard, nobody bothered to run a wire to the house or for that matter wire the house for power. So it was, when we went to bed in the 2 rooms that were "upstairs" we climbed the steps which were painted white but have a black mat on each step so there was pretty much no light other then that small flickering lamp that mom had lite in the kitchen below, that being a kerosene lamp. We would go up-stairs using all 4 limbs, hands on the step above and feet on those below till we reached the top which had no landing to speak of, just turn and you were in the first upstairs room, there was a drop down door that we had to lower once we were up there to keep the heat in for the downstairs, In the second room were two beds, one large for my sisters, and one small for me. That room had a lover that could be opened to allow heat from the coal space heater in the living room where my folks slept in winter to stay warm. The upstairs had no finish on the walls or ceiling, in the winter the shingle nails on the roof would have heavy frost on them, and when we woke up it was common to see your breath in the morning light that came from the one window that faced the East. It was a good life though, we all grew up to be successful, and all of us were very creative, you see you had to be because there were no televisions, phones just a battery pack radio that was high enough that we could not reach it to turn it on without a stool to stand on (batteries were expensive in the 50's)
An addressable led strip with a ESP32 with WLED and two IR sensors (one on top of the staircase and one at the bottom) could do the trick relatively easy (even change the effect/brightness at certain times of day). Maybe even coupled with home assistent to intergrade in the smart home.
I think you should use higher voltage such as 12 volt with an optocoupler as low voltage is easily get affected by noise, and also use a 555 timer ic for PWM and time delay which can be triggered by a normal IR sensor. No need for a attiny. But all and all It was a very nice video for me because it gave me a chance to think.
Thanks for sharing your problems so others don't "fall" into the them. Engineering is all about iterating so no shame in prototyping. Can I suggest having LIN bus or another com line among all the sections so not only will they light up where you step but where you are going to step. The LIN bus would also help you debug so you only have to talk the primary chip to debug the entire strip.
Well, I spent so much time on this project that I kind of had to make it into a video. And no, coding is not the problem because everything worked fine when testing with less units. I think EMC is the big problem here.
THANK YOU SO MUCH for sharing your entire process - it is so refreshing to see setbacks and someone holding their hands up - too many people think experts know everything -crazily - this is the EXACT thing ive been thinking about doing - I'm so glad its not just me that gets stuck and frustrated so much you have to just walk away and come back with a fresh mind later. Thank you for sharing this - I would be VERY interested in the solution you come to when you do.
I was going to make animated lights on my staircase but decided a simple "all on", when a person is detected, was safer. I used a multivibrator circuit to turn on the lights for 30 seconds, and re-trigger whenever motion is detected. Fancy lighting effects look cool, but are distracting and dangerous IMHO. I love your method for mounting the LEDs. I chose to do the massive wiring to each LED approach, lol.
Yea that's the I way. A local shop to me installed one of those sequencial stairs that was sequential in both the steps and the sides (not shure if it makes sense), and I found it super distracting going up those stairs until I sort of memorised them and got to go up without caring about the light.
A bunch of sensors can mess with each other. I went with radar (Doppler GHz) at top and bottom, they are far enough to not see each other. Then you can turn on all the LEDs in sequence (away from motion) and control them all from nice microcontroller. Giving you tad more time to move past them. Then shut them down again in sequence. And those radar sensors are so awesome!! Oh, addressable LED’s could share power and data. And I agree, we love learning along side you. My projects go the same way! ;)
You simply should enable only 1 distance sensor at a time, not all at once. Distance sensors are interfering with each other when they are simultaneously firing . Also most probably the distance sensors are the ones that create the noise.
But how do you activate only 1 at the time? To know if person is standing on the step you need for distance sensor to be turned on at all time and you have no idea from which direction up or down someone will go
This reminds me of a project I built back when I first started tinkering with electronics: A motion sensor based light controller with a PIR motion sensor. It worked beautifully most of the time, but whenever someone would flip on the light in the next room over, it would trigger the motion sensor resulting in the lights coming on in the middle of the night. I suspect I had similar issues with the input voltage fluctuating too much and confusing the sensor. Thank you for the great video, keep up the good work and don't give up!
@@greatscottlab perhaps it might help to reduce the size of them, and for the ground plane to only really extend under the attiny and the LED controller IC. It might also help to run VCC and ground over an external (twisted?) wire pair, but that would somewhat defeat the purpose.
If not already done, it can be beneficial to connect the aluminum profile to GND and/or protective earth. This can reduce the "noise" fields by giving those a decent path to "conduct" to.
I love the I tried it this way and it failed, and tried it another way and it failed over and over relentlessly. the break is going to help you piece together all those moments and you're probably going to discover the cause: whether it's some frequency, heavy pulldown, floating grounds etc. Kudos "Great Scott" for the endeavours.
I'd recommend using the table saw to open the groove in the wood. Even without a wide blade (dado blade, not available in Europe), doing multiple cuts will be far faster and easier than using the router
And this is especially a good way to get the outer cuts straight. Alternatively a super simple router table could be made using a plate ans some scrap wood (if one have a router but no table saw).
100% need the next iteration of these and update video! Been wanting to something like this for a long time. I did a much "older" variant using lasers and LDRs on each step as well as some ultrasonic stuff -- this is a much nicer solution once the bugs get ironed out
when the location is less important than the detection, you could add two sensors, one at the front of the PCB and one at the back, and add an AND or an OR gate for the signal (depending on the noise and/or functionality you want). That also solves one issue your current setup has: your lights switch on when you technically already stepped forward instead of "leading the way".
I feel your pain, my friend… we all have those frustrating project experiences. Thank you, for sharing yours. I’ll look at the files and think about this and send any thoughts later. Great video, as always! 👍😊👍
It's immensely relieving for me to see that I'm not the only one who doesn't get it right immediately. Thank you GreatScott! And: would love to hear how you'll fix this (and I know you will!)
Great project. I love the fact that you take us all through the design, build and problem solving and not just "look I build X and it's working". I can't wait to see how V2 gonna be.
You could use your clear covers and add defusion to them and leave the area in front of the sensor clear so they still work as intended. Projects like this rearly work right off so it's nice to see a realistic expectation here where it doesn't work right the first time. I like your solution for mounting the light strips and not attaching to the staircase itself you even got a good colour match, well done.
This has got to be a frustrating project. It takes humility to admit you have failed. I look forward to the sequel video where you finally solve the problem and provide a valuable lesson learned to the rest of us plebes.
My spontaneous low tech contribution. Stay Analog: Replace the sensors with strain gauges under the stairs. Use some OPV signal conditioning (low pass + gain + some integration) and trigger the LEDs from that. You would send a dampened variant of the signal to the neighboring stripes and mix to their signal conditioner for the "light Aura" effect. Attaching strain gauges below the stairs increases the unwanted glue-on-the-furniture, but damn, this thing would be cheap and work forever due to its simplicity! It even distinguishes pets from human, if you trim the gain stage. Pets tend to have better eyes than the median tech-nerd!
Addressable / Zoning (with few staircase steps), based on dimensions e.g 1-5 steps ~ Zone 1, 6-10 ~ Zone 2 and so on, this will provide fault tolerance too, not all LED strips will go bad at any given point in time and thus some visibility will always be there to prevent a fall from staircase.
The IR sensors might break eachother if they power at the same time, you should probably do some multiplexing to make it cycle through the sensors (or at least not power the ones directly besides them all the time
It's good to see a failed/incomplete project as most hobbyists will struggle through similar things. Looking forward to part 2. Thinking outside the box, It's possible you could control the LED driver with a voltage signal instead of PMW which might lower your noise? Another thing, which may not matter, the capacitors on the power regulators inputs are all in parallel. The led driver chip could be replaced with a mosfet driving the LED and the Atiny could control dimming, this would then remove the need for a 12v>5V just run the whole strip on 5 & 3.3V
I love it, amazing project. No matter if it was a fail. Thank you for show us that sometimes something could be wrong. I enjoyed a lot with the concept, process and the creation. My wife also sit next to me often and say to me: Do it but don't broke it 😅 I'm looking forward for the V2 and probably you should consider to sell it! Could you consider to use a white cover for the Ledstrips and a small transparent section for the sensors. Great job!! 👍
I had a project like this but in the end i used a commom led strip, cut it in the minor lenght (3 LEDS) per step, fixed it in the handrail facing downward and used a PIR sensor to turn on the hole stair case. Also its integrated in Homeassistant so it can serve as indirect light for the living room. It's not that fancy indeed, but helps a lot
You can easily make this circuit with using only 3 transistors as a switch, One transistor for onboard sensor, and the other two for previous and next sensor signal and can be easily Daisy chained. Transistors will turn the lights on and off. And dimming can be controlled by using led driver. Hence, no need for microcontroller
I made similar lights. However, I took another approach. I use PIR sensors (trimmed physically to a narrow slot) as start and stop gates at the bottom and top of the stairs then I time the stair animation based on the average pass between these points. Yes I don't detect individual steps but you'll notice that in exception of running, each staircase has a very consistent traversing speed. I was able to easily pick the speed of the lights switching ahead of me with me early overtaking the strip. Best bit? I used £20 worth of components for 11 steps.
I'd like some lights like this for my sailboat's companionway. Just to light up the ladder for entry/exit. Having them light up with human presence would save on battery usage and give the perfect amount of light without destroying night vision. I'm going to be living aboard so lights like this would be very handy for the companionway and possibly in the head as well. Thanks for the great idea that's giving me some good inspiration!
Had very similar idea back in 2021 while building my stairs. Wanted to do all just using analog discrete parts. Never actually got to it. Good luck with your project and thanks a lot for the kick I probably needed to get back to this project.
I honestly don't consider this project a failure. The majority of the staircase seems to work passably well, and the lights that are stuck on, are at the top of the stairs providing you an unintentional safety bonus of always knowing where the first step down is. Sure, it's not by design, but I'd be happy with this.
Try using the RCWL-0516 microwave motion sensors. They work incredibly well and can be used behind solid objects so using your frosted diffuser would be no problem. The distance can be adjusted with small variable resistors to fine tune. Additionally you could add some wifi (esp8266? Or similar) and connect to home assistant to get motion alerts or trigger lights to blink in your bedroom or office so as to notify you when someone is coming up the stairs.
As was already pointed out, you definitely need one diode on each "OUT" or one diode on each "IN", otherwise you're just connecting everything together and it can't possibly work as intended. Apart from that, make sure the distance sensors aren't interfering with each other (by seeing each others IR scattering of the other side of the stairs) and make sure you only evaluate the light sensor measurements when the LEDs are off, so they don't measure their own light. I can't really imagine noise from the switching regulators being that bad, assuming they're laid out decently. The voltage drop across the PCBs also seems alright, given the fairly even brightness across the length of the strip. So I would expect the power supply to be good/clean enough.
It took me three weeks four different tries with the programing (worse than you in that area). Finally went to a friend about the code he showed me that using a state machine was the best case and it mostly works. The issues on the board like noise and signal integrity were mostly the board layout and lack of proper filtering. I have built this in two different places, and for the most part they are still working. Had an LED go out on one about a year back. Mine are made to react to the step before being on or off state. Also had to angle the motion sensor up at a thirty five degree angle to keep the cats from playing with the lights. Peace
This is great, Scott! As other comments mentioned, I would consider using addressable led strips, but also taking advantage of the different colors you can turn on. At night, the blue light is the one you should avoid, so you might turn on the leds on yellow color.
Using a linear regulator to power the LED would be a good idea because it's not a linear dipole. Since their voltage is roughly constant in most the the upper part of the current supply that means that the regulator won't get much voltage until the current gets quite low. And since P=U.I and you either have low voltage or low current going on your regulator, it will be relatively efficient.
I think I have a few suggestions for this, (A note before we begin, I'm not any working professional and have no experience. Just some interest in electronics) 1. Add some heavy capacitors(~100-1000 mF) directly to LEDs and a way so limited current flows(resistor .5-5k) to them LEDs and capacitors. That way, whenever led turns on or off, they happen smoothly. 2. Add 5 pins, 2 big on the sides for power and 1 in centre for programming the microcontroller(as used in the video), all at once. 3. For the remaining 2 pins, they'll be 'in' and 'out' data pins located between the centre and power pins. Use them for Daisy chaining.It'll be like this: (+)-------------------------------------------------------(+) (in)-------------through the MC------------(out) (prog.)---------------------------------------------------() (out)----------through the MC---------------(in) (-)-------------------------------------------------------(-) i. When a board senses something in the proximity, it'll obviously turn on but also send the signal to both ends(1 for on, 0 for off). ii. When a board receives the signal from either end, it'll turn on the led but won't pass the signal to output pins. This way you can stand on any step and the next one will be ready for you to see and it's simple to implement even without the microcontroller. If you want multiple steps to light up, use frequency/data method. In this method, board with proximity will generate and pass a certain frequency to both ends and receiving board will turn on but will pass on a different frequency to both ends and so on upto the frequency that it won't be passed anymore by the receiving board. Now you can light up n number of stairs with n number of frequencies. Easier said than done, this frequency/data method will have few problems of its own like calibration, noise, input-output loop, etc. Edit: A new method came to my mind, remove those 2 in and out pins and add a common frequency rail where each MC will generate their own frequency, and also they will always monitor the frequencies and know exactly on which step you are and respond accordingly by turning on. This method is quiet power inefficient so add a sleep/low power mode where it won't monitor frequencies if there's already a certain amount of light on stairs. It's getting long to explain so I'll stop here. One more thing, you don't need boards that long, it can be fitted in 2-4 inch length. You can use wires to chain them. I came across the ad of such products a few months back but like you said I also didn't find them good enough for my liking (idk what's their state now). Hope to see you come back to this soon, take care. 😉
If you're communicating with each board. You can send out a simple int up and down range of the triggered board. So for example sending out 3. This would be how many boards are lit up/down stream. The next board sees the 3, turns on, changes 3 to 2, and passes on the number. The next board lights up, turns on changes 2 to 1. The next board lights up, changes 1 to 0. The next board gets zero and does nothing. Also depending on how you trigger the board, itself being triggered vs being triggered by a neighbour will give you a direction and you can light up boards in only that direction.
Perhaps use your communication lines to interlock the operation, based on whether the top or bottom step is detected first, then selectively enable LED strips as you walk up, while disabling strips after they're passed. So, for 14 steps, only step 1 and step 14 would be actively watching, with not more than one adjoining step standing by: 1. When proximity is detected at 14, step 1 is disabled, and step 13 enabled for proximity detection. 2. When proximity is detected at 13, step 12 is enabled for proximity detection. 3. When proximity is detected at step 12, step 11 is enabled, but step 14 is disabled and the lights are allowed to time out. 4. Go back to step 3, decrementing numbers down the steps, until the bottom is reached. 5. When the stairwell is clear, go back to watching just step 1 and 14. Of course, this doesn't allow for the possibility of someone coming up the stairs at the same time, or anyone who likes to skip stairs. I'd expect neither is typical of your nighttime use case, though.
I LOVE THIS PROJECT!!! I know you feel frustrated by it but you definitely when through all the troubles and troubleshooting problems someone would have gone through when developing a small project and i love it!!!
Now that I've seen your results, I'd probably add communications between adjacent strips. Then one board could tip off the neighboring board to go into low light mode for a bit, or enable a timed cascading On/Off sequence for a more appealing presentation and integrated design. I'm eager to see the final design!
Just one thing... You have an IR sensor there and LEDs which tend to heat up pretty quickly.... Combine the two to get a perfect recipe for a disaster. P.S. I would definitely go with an ultrasonic distance sensor. Waaaaaaaaaaaay more reliable thing.
I have done something similar. Use a radar sensor (the cheap microwave I think frequency ones) and you can route a hole in the wood and mount it from the backside so you don't see it, they work through wood. Have one at the top of the stairs, one at the bottom. Build one control board to manage those sensors, and to turn the LEDs on and off. Run wires to the LED locations, and this can all be done on the backside of the wood so all you'll have on the front of your nice wood strip is the hole with the LEDs and you can add disfussion covers easily. Code it to turn the LEDs on in ascending or descending order (depending which sensor triggers) over 1-2 seconds, using PWM, and auto-shutting off in the reverse order after 20 seconds or so.
yeah I have the same problem with see through stairs. I had other ideas but seems we may move out potentially, so we'll see 6:50 just drill 2 small holes. makes it very directional and not false trigger
Using multiple values of decoupling capacitors for instance 100pf, 10pf, and 1pf may help filter the noise as pwm will generate noise over a lot of different frequencies
Love the overengineeredness! If i would make this however id keep the simple stuff simple as the hard stuff is hard enough. Have you considered a single microcontrollwr for the whole staircase? Even an esp32 suffices. Add an ial lightstrip like a normal 28xx 12v one (12 for the large distance and voltage drop) and a seperate line with distance sensors. When all the sensors and lights are on 1 microcontroller, your code is immensly simplified. Just a maxvalue filter on the distance for the currently steppen on staircase and a minimum threshold. You can even normalise the noise by averaging out all the sensor input. This is the way
You can use diffuser if you can raise proximity sensors and make holes in the diffuser so that the diffuser cannot mess with values and off course it will look way cleaner and professional
you could make a digital line between the neighbouring lamps. Each could tell to their direct neighbours (i-1 and i+1) that he is detecting you, so the next stair can light up, and also this could serve as one more layer to filter out random lights flashing. Those long pcbs can handle one or two more traces..
Also consider large programming / debugging pins. If you space the distance sensor and the leds you can alternate frosted and clear cover so,that the lights look nicer. Or even only the area of the distance sensor clear / with a peeking hole
You can try using one master PCB with a microcontroller chip and the other PCBs to just have power lines with LEDs and proximity sensors.🙂 Or just use just two sensors at the first and the last step to light up as how WS2812 leds light up with a drop effect.🙂
When you make a long-run data transfers, you have to make sure you sorround the data lines with a lots of ground and keep the noisy 12v away. (And use dedicated pulldown resistors at the input side with a small capacitor on the output side of the data lanes.) 2nd suggestion is to only pull the output high if a feet is detected, otherwise pull it down. With a small delay on each side(after input interrupt and output sending)
You can buy dimmed cover and undimmed cover, and make mozaic from them, to have undimmed above sensor and dimmed above LEDs. It means, that you will have cover from many small covers, which is firstly bad, but, if you must to repair some module, you can uncover only that part. And also, I will do one PCB to fit into rail in aluminium profile, and one PCB above this, connected via headers, to have posibility to remove/repair/replace it without pulling out any PCBs from rail.
I really like your idea of automatic lights for stairs. Thinking about design aspects for safe passage observing you walking the stairs. Is the IR sensor in the middle of the board or at one end? does it have sufficient spread angle to detect you soon enough on top and bottom step regardless of the side of the step you walk on? As for the lighing logic, could group the lighting sequence, so regardless of walking direction, leds light for the current and next step at 100% plus 20% for the third step. Giving light before you get there, otherwise fast walking would not give time to see properly - a recipe for an accident. For your power issues, maybe just try a 78L05/78L03 for each PIC supply and use PWM for LED brightness. You could have one master board with two pcb mount trimpots, one for low brightness and one for full brightness. Giving you flexability without reprogramming. Then theres the question of what should the lights do if you stop on the stairs. I wonder if theres a different diffuser type that allows IR through???
I have that same tiny hotplate. I cut a couple of blocks out of wood the same height as the hotplate to help support larger boards. It's a lot safer to use!
Use the diffused covers, but drill a small hole right above the distance sensor. You might even get away with a few smaller holes in a triangle or square pattern.
It sounds like you will be needing new PCBs for this project. Since you cannot use the diffuser lenses, maybe consider getting your new PCBs made in white with only the LEDs facing the stairs and all other SMCs on the wall side of the PCBs. The LEDs will blend into the white PCBs better and the strip would have more of a diffused look when the LEDs are off. Great job so far. Looking forward to the final product!
What you're trying to do is very neat. But it looks challenging with the problems you are facing. A while back I developed an ESP32 based controller for an LED strip that sits on the bottom of a shelf over our toilet. It uses an ultrasonic distance sensor (HC-SR04) to read the distance. I put my hand under the sensor and the LED strip lights up for a certain amount of time (usually 3 minutes). This way I can use the bathroom at night without killing my night vision and waking myself fully up. I thought about the staircase application you are trying. Rather than placing a sensor at intervals along the strip I figured I'd just use my existing strip controller at one end. I'd put another standalone controller at the other end and transmit the signal that someone passed the sensor using ESP-NOW to my existing controller to also activate the strip. If you wanted to simulate someone walking you could light up the segments of the strip using timing. But that just seems unnecessarily complicated. I'd just light up the entire strip for a preset amount of time.
That is such an excellent project! I see where you were headed with it, (and kudos for working as a carpenter for a bit- THAT can be quite frustrating if it's not something you do all the time!) I learned some things, and JLCPCB pricing info... I can't wait to get home so I can work on some original design boards to try them out! Don't feel bad, that try, fail, retry, walk away loop happens to all of us!
you can use wled,they have a mode that you can attatch ws2812 addressable led and will light up one by one with ultrasonic distance sensor,designed exactly for this use
Socket the connection between boards so you can separate then and take them back to the bench to work on, soldering them all together into one giant circuit is just asking for problems. If you were to re-design I'd probably build small boards that sit behind a WS2812 strip, and utilize the existing WS2812 circuitry for LED control and just implement the proximity sensing on that board. If you want bidirectional messages, get two of the narrower strips and run one in each direction and catch the signal from the upstream and downstream board. Only re-invent the parts you NEED to, not the whole thing.
